{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:7N4SRAPHW5CH7GCKIPOD55SJL2","short_pith_number":"pith:7N4SRAPH","schema_version":"1.0","canonical_sha256":"fb792881e7b7447f984a43dc3ef6495e8aa412832aac15cf9e5bb710ac634417","source":{"kind":"arxiv","id":"2608.02046","version":1},"attestation_state":"computed","paper":{"title":"CompanionBench: A Theory-Anchored, Real-World-Grounded Benchmark for AI Emotional Companionship","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Guangjia Chai, Jihao Huang, Junchen Wan, Lei Wang, Yao Liu, Yuming Huang","submitted_at":"2026-08-03T10:44:04Z","abstract_excerpt":"LLM companions are deployed at scale in personally consequential settings, yet poorly evaluated. Existing benchmarks use hand-authored scenarios and prompted simulators, aggregate empathy into one score, and overlook judge biases such as same-family favoritism and scale drift. We introduce CompanionBench, an interactive bilingual benchmark. To our knowledge, it is the first companion benchmark to ground both its scenarios and a trained user simulator in de-identified real-world data. A hidden disclosure gate branches each persona's trajectory on the agent's own behavior, controlling the intera"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2608.02046","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-03T10:44:04Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"1245c0b0a90e1c51efd62ac7aa9f1078c5c56000054ebce2c2d494d874fc064a","abstract_canon_sha256":"735d63628c238f74cf1822ba523ea08bc510958ee0cd2067243d3d00576c9fe2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-04T02:10:44.145072Z","signature_b64":"56seaux4RROpzQNGUzxLS+wwqIlVlkVHvlaCtmMtnRAyqZ5gI3rQzvdOrDc0aSBIADEZkPrm5CEjmzBtrp79Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fb792881e7b7447f984a43dc3ef6495e8aa412832aac15cf9e5bb710ac634417","last_reissued_at":"2026-08-04T02:10:44.143625Z","signature_status":"signed_v1","first_computed_at":"2026-08-04T02:10:44.143625Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CompanionBench: A Theory-Anchored, Real-World-Grounded Benchmark for AI Emotional Companionship","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Guangjia Chai, Jihao Huang, Junchen Wan, Lei Wang, Yao Liu, Yuming Huang","submitted_at":"2026-08-03T10:44:04Z","abstract_excerpt":"LLM companions are deployed at scale in personally consequential settings, yet poorly evaluated. Existing benchmarks use hand-authored scenarios and prompted simulators, aggregate empathy into one score, and overlook judge biases such as same-family favoritism and scale drift. We introduce CompanionBench, an interactive bilingual benchmark. To our knowledge, it is the first companion benchmark to ground both its scenarios and a trained user simulator in de-identified real-world data. A hidden disclosure gate branches each persona's trajectory on the agent's own behavior, controlling the intera"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.02046","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2608.02046/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2608.02046","created_at":"2026-08-04T02:10:44.145119+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.02046v1","created_at":"2026-08-04T02:10:44.145119+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.02046","created_at":"2026-08-04T02:10:44.145119+00:00"},{"alias_kind":"pith_short_12","alias_value":"7N4SRAPHW5CH","created_at":"2026-08-04T02:10:44.145119+00:00"},{"alias_kind":"pith_short_16","alias_value":"7N4SRAPHW5CH7GCK","created_at":"2026-08-04T02:10:44.145119+00:00"},{"alias_kind":"pith_short_8","alias_value":"7N4SRAPH","created_at":"2026-08-04T02:10:44.145119+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7N4SRAPHW5CH7GCKIPOD55SJL2","json":"https://pith.science/pith/7N4SRAPHW5CH7GCKIPOD55SJL2.json","graph_json":"https://pith.science/api/pith-number/7N4SRAPHW5CH7GCKIPOD55SJL2/graph.json","events_json":"https://pith.science/api/pith-number/7N4SRAPHW5CH7GCKIPOD55SJL2/events.json","paper":"https://pith.science/paper/7N4SRAPH"},"agent_actions":{"view_html":"https://pith.science/pith/7N4SRAPHW5CH7GCKIPOD55SJL2","download_json":"https://pith.science/pith/7N4SRAPHW5CH7GCKIPOD55SJL2.json","view_paper":"https://pith.science/paper/7N4SRAPH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.02046&json=true","fetch_graph":"https://pith.science/api/pith-number/7N4SRAPHW5CH7GCKIPOD55SJL2/graph.json","fetch_events":"https://pith.science/api/pith-number/7N4SRAPHW5CH7GCKIPOD55SJL2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7N4SRAPHW5CH7GCKIPOD55SJL2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7N4SRAPHW5CH7GCKIPOD55SJL2/action/storage_attestation","attest_author":"https://pith.science/pith/7N4SRAPHW5CH7GCKIPOD55SJL2/action/author_attestation","sign_citation":"https://pith.science/pith/7N4SRAPHW5CH7GCKIPOD55SJL2/action/citation_signature","submit_replication":"https://pith.science/pith/7N4SRAPHW5CH7GCKIPOD55SJL2/action/replication_record"}},"created_at":"2026-08-04T02:10:44.145119+00:00","updated_at":"2026-08-04T02:10:44.145119+00:00"}